Method to assign upstream timeslots and codes to a network terminal and medium access controller to perform such a method

Abstract

A method to assign upstream timeslots and codes to a network terminal in a point to multipoint communications network includes a step of assigning at least one code to said network terminal for inclusion within a network terminal grant assigning upstream timeslots to said network terminal. A medium access controller for performing said method is as well described. The network terminal is adapted to extract said at least one code from said network terminal grant, and to encode upstream data packets with said at least one code before upstream transmission of the encoded data packets, upon receipt of this network terminal grant.

Description

[0001] The present invention relates to a method to assign upstream timeslots and codes to a network terminal as is further described in the preamble of claim 1, to a medium access controller to perform such a method, which has the features as is described in the preamble of claim 15, and to a network terminal as is further described in the preamble of claim 29.

[0002] Such a method, medium access controller and network terminal are already known in the art, e.g. from EP specification 0544 975. Therein a time division multiple access, hereafter abbreviated with TDMA, system is described wherein individual network terminals are assigned upstream timeslots for transmission of upstream data packets to a central station, by means of network terminal grants generated by a medium access controller incorporated within the communications network.

[0003] This method and system is however one-dimensional, taking only the time-division aspect into account. There is nothing mentioned how to generate a medium access control method for two-dimensional systems, where for instance both TDMA and code division multiple access, hereafter abbreviated with CDMA, are combined.

[0004] U.S. Pat. No. 5,894,473 describes a multiple access communications system and method using code and time division. The method comprises coding information signals with CDMA codewords to be transmitted over a common frequency spectrum, time compressing the CDMA code words for transmission only during allocated timeslots, activating a receiver only during the allocated timeslots to receive and decompress the time compressed CDMA codewords and decoding the decompressed CDMA codewords to recover the information signals. The allocation of the codes takes place locally in the network terminals themselves, for instance on the basis of physical level parameters such as the power level of the to be transmitted signals. Reference is also made to another U.S. Pat. No. 5,844,894 indicating that independent channel assignment strategies can be applied in different timeslots. These strategies are however used between two base stations, corresponding to two central stations in the present invention, and not between the base station and the radio terminals themselves as is the subject of the present invention. Moreover, in the referenced US patents timing synchronisation between the central station and the terminals is very important, as is reflected by the availability of the burst timing controller. There is also no indication within both US patents how to centrally allocate these timeslots.

[0005] An object of the present invention is therefore to provide a method and a medium access controller adapted to simultaneously and centrally allocate timeslots and codes to the user network stations within such a two-dimensional combined TDMA/CDMA communications network and capable to handle different degrees of synchronisation between the central station and the user network terminal.

[0006] According to the invention, this object is achieved due to the fact that said method includes the steps as described in the characteristic portion of the first claim, that said medium access controller is further adapted as is described in the characteristic portion of claim 15, and that said network terminal is further adapted as described in the characteristic portion of claim 29.

[0007] In this way, a generic two-dimensional medium access control protocol is provided wherein both the time slots as well as the codes are centrally assigned by the medium access controller to the individual network terminals, by means of a dedicated medium access control mechanism which is centralised within the network and for which a dedicated medium access control layer is used. Such a process of central assignment of timeslots allows much looser requirements of synchronisation as will also become more clear from the descriptive part of this document.

[0008] Another characteristic feature of the present invention is that said method further includes the steps as described in claim 2 and that said medium access controller is further adapted as is described in claim 16.

[0009] In this way the two-dimensional constraints related to the physical interferences between the network terminals, as well as between the individual communication channels are taken into account. This is extremely important for CDMA systems where these physical interferences limit the use of available codes. These physical interferences may include Rayleigh fading, and multi-user interferences.

[0010] Another characteristic feature of the present invention is that said method includes the steps as described in claim 3 and that said medium access controller is further adapted as is described in claim 17.

[0011] By taking into account the total load within the network, for allocating upstream timeslots and codes, efficiency and throughput are optimised.

[0012] As described in claims 4 and 18, by letting this load to be dependent on individual requests transmitted by the terminals, these requests being indicative of the amount of upstream data packets the network terminals intend to transmit, a simple and effective method for determining the load is provided.

[0013] Still another characteristic feature of the present invention is described in claims 5 and 19.

[0014] When the time and code allocation is dependent on a delay parameter, a compromise between throughput within the network and delay requirements pertaining to individual connections is obtained.

[0015] This delay parameter can as well be based upon the requests transmitted by the individual network terminals, or it may be generated from the connection admission control parameters, as will become clear from the descriptive part of this document.

[0016] Yet another characteristic feature of the present invention is described in claims 6 and 20.

[0017] Thereby a selection is made from a plurality of code allocation procedures. This selection and subsequent performance of the selected procedure on one hand allows to remedy specific problems within the network, while at the same time aiming to compromise parameters as throughput and fairness.

[0018] Further characteristic features of the present invention are described in claims 7 to 10 and 21 to 24.

[0019] These claims describe specific ones of these time and code allocation procedures. A first procedure is focussed on optimising throughput by reducing the load within the network, another is focussed one on fairness, thereby allowing as much as possible network terminals to be served simultaneously, a third one represents a compromise between the previous two, and a fourth one aims at guaranteeing delay boundaries for instance in case of guaranteed constant bit rate connections. In any of these procedures, as was already mentioned by the features described in the second claim, the physical boundary conditions are taken into account.

[0020] Still another characteristic feature is described in claims 11, 12 and 25 and 26.

[0021] In case some network terminals intend to transmit upstream data packets pertaining to connections for which some connection admission control parameters were negotiated during connection set-up, the second and third procedures thereby allow to take these connection admission control parameters into account. The total result is that a generic medium access control method is provided taking into account a maximum of parameters related to the physical medium, the load within the network, delay constraints and quality of service parameters.

[0022] Still another characteristic feature of the present invention is described in claims 13 and 27.

[0023] By executing the method at predetermined time intervals, which in general correspond to the TDMA timeslots, at each timeslot a multipermit or multigrant is generated whereby thus one or more terminals are assigned one or more codes for subsequent upstream transmission of data packets to the central station.

[0024] Still a further characteristic feature of the present invention is described in claims 14, 28 and 30.

[0025] The method is thereby not only applicable for generating grants per network terminal, but allows to generate multipermits per service category queues per terminal.

[0026] The above and other objects and features of the invention will become more apparent and the invention itself will be best understood by referring to the following description of an embodiment taken in conjunction with the accompanying drawings wherein:

[0027] FIG. 1 represents a communication network according to the invention, and

[0028] FIG. 2 represents the central station of the communications network of FIG. 1, including a medium access controller according to the invention.

[0029] FIG. 3 represents a network terminal of the communications network of FIG. 1, according to the invention.

[0030] The communications network of FIG. 1 is composed of a central station CS and a first plurality of network terminals T1, . . . , Ti, . . . to Tn. The central station is coupled to these network terminals via the cascade connection of a common transmission channel, for instance a copper link for a power line communications, abbreviated with PLC, network, and respective individual network terminal channels L1, . . . ,Li, . . . ,Ln, also for instance consisting of copper cables in the case of the aforementioned PLC network. In other environments, for instance in UMTS networks which stands for Universal Mobile Telephony Services networks, the common transmission and individual channels consist of radio links. The network hence has a point-to-multipoint architecture in the downstream direction, which is the direction from the central station CS to the network terminals T1 to Tn, and a multipoint-to-point architecture in the upstream direction, i.e. the direction from the network terminals T1 to Tn towards the central station CS.

[0031] Networks having such a point-to-multipoint architecture in the downstream direction are for instance the already mentioned PLC and UMTS networks, but also satellite networks.

[0032] In the downstream direction, the central station CS broadcasts information to all network terminals T1 to Tn. In PLC networks this information is for instance empacked in so-called downstream frames. In the opposite direction, the network terminals T1 to Tn commonly share the common transmission channel in a time-multiplexed way. This means that different network terminals transmit information to the central station CS in different timeslots. Each network terminal thus sends upstream information in short bursts to the central station. The upstream timeslots constitute so-called upstream frames for the aforementioned examples of UMTS and PLC networks.

[0033] To be allowed to send a burst in an upstream timeslot, a network terminal, for instance Ti, has to receive a permission or network terminal grant from medium access controller MAC, usually included within the central station CS, as is also drawn in FIG. 2.

[0034] At regular time intervals, such permissions are downstream broadcasted by this medium access controller, for instance by means of a dedicated cell. In these dedicated cells the contents of grant fields precisely define which network terminal is allowed to occupy which upstream timeslot. To however allow to increase the amount of users or network terminals within these networks, the transport channel is further divided into sub-channels according to orthogonal codes. This results in a combined TDMA/CDMA access mechanism which can be used for the aforementioned UMTS, PLC and satellite networks. This means that for instance the same frequency band for transmission can be shared by a multitude of network terminals by coding the information such that transmitter and receiver can only discriminate the data on the basis of the code. As is however also well known in the art, the orthogonality is not always guaranteed, especially in the presence of Doppler effect, terminal interference etc., where data coded differently is corrupted and misinterpreted as originating from another terminal.

[0035] Traditionally these codes are allocated for the duration of the communication itself. Due to the aforementioned interference effects, the amount of codes which are really allocatable is however limited and smaller than the amount of codes within the theoretically available second plurality of codes. By however allocating the codes per timeslot, thereby taking these physical interference effects as well as the instanteneous load and delay within the network, into account at each of these timeslots, a more optimised code allocation procedure is obtained. This itself results in a more efficient usage of the transmission medium.

[0036] The present invention therefore concerns a method for generating and assigning a network terminal grant associated to an individual network terminal, for instance network terminal Ti, part of the first plurality of network terminals T1 to Tn, whereby at the some time also one or more codes are allocated to this network terminal. The present invention thereby concerns as well a medium access controller MAC which is adapted for executing this method. This medium access controller has central functions and is as such in most cases incorporated in the central station. However it may also be a stand-alone unit residing somewhere centrally within the network.

[0037] A detailed embodiment of such a medium access controller is schematically depicted in FIG. 2.

[0038] This medium access controller MAC includes a load monitoring means, denoted LMM adapted to determine an activity parameter denoted L and which is related to the load within the network. The term load is thereby to be understood as related to the amount of upstream data traffic individual network terminals intend to transmit. There are several possibilities to determine this load, one method thereby being a predictive method such as is described in the non-published European Patent application 99401051.0, filed by the same applicant. Another method, as is depicted in FIG. 2, uses upstream requests earlier transmitted upstream by the individual network terminals to the medium access controller, and including information related to the amount of these upstream data packets these individual network terminals intend to transmit. This information can for instance be the amount of data packets queued within these terminals, or may indicate that a certain threshold within such a queue is surpassed. These requests are denoted R1, . . . , Ri to Rn and respectively indicate requests transmitted by T1, . . . , Ti to Tn. The activity parameter L can than be calculated from these requests in different ways, for instance by counting together the amount of data packets, in case the requests indicate so. Other possibilities for calculating L can consist of counting these requests indicating a surpassing of the threshold.

[0039] The medium access controller MAC also includes a physical constraint means, denoted PCI, and adapted to determine from the available codes within the second plurality of codes, from parameters related to the physical interference between the terminals and the individual network terminal channels, a set of two-dimensional constraints. These constraints are in fact a set of mathematical expressions, to be considered as boundary conditions for the calculations further performed within other blocks in the medium access controller, as will be described in further paragraphs. The parameters related to the physical interferences between the terminals and the individual network terminal channels thereby are such that these also incorporate the degree of synchronisation between the network terminals and the central station. In this respect the method can thus handle different degrees of synchronisation since this will be represented by different values of the resulting set of two-dimensional constraints. The physical interference input parameters are in general stored within a memory in the PCI device itself, as well as the codes itself. The set of two-dimensional constraints generated by the PCI device, is denoted Phy.

[0040] The medium access controller MAC further includes a jitter determining means, denoted JDM. This jitter determining means is adapted to determine a delay parameter, relative to delay at some delay-sensitive connections within the network. Delay is only relevant to these connections with strict delay-bounds, such as the constant bit rate connections, for which strict delay parameters were initially agreed during connection set-up. The delay parameter J is therefore calculated from traffic information only relative to these delay-sensitive connections. In case the network terminals have included the information with respect to these delay-sensitive connections within their request fields, the requests are also input parameters to the jitter determining means JDM. In case, such as is depicted in FIG. 2, information with respect to these delay sensitive connections, for instance the constant bit rate connections, within the terminals, is not incorporated within the request fields, first a kind of pseudo constant bit rate requests are calculated from these specific connection admission control parameters, which are then consecutively transmitted to the jitter determining means which calculates the delay parameter denoted J from these pseudo constant bit rate requests Rcbr.

[0041] The pseudo constant bit rate requests Rcbr are generated within the request table, denoted RT, which is also incorporated in the medium access controller MAC. This request table temporarily stores all incoming requests, and also generates the pseudo constant bit rate requests from traffic and connection parameters pertaining to these constant bit rate connections. These traffic and connection parameters are centrally stored within the central station CS within a connection admission control memory, denoted CACM. This memory is continuously updated each time new connections are set-up, and contains the traffic and connection parameters pertaining to each of these connections. For the constant bit rate connections, the traffic and connection parameters are schematically denoted Tcbr in FIG. 2. The request table RT then is adapted to receive these Tcbr parameters and to generate therefrom the pseudo constant bit rate requests Rcbr. A possible way of doing so may consist of assigning a counter per constant bit rate connection, initialising this counter with the desired inter-arrival time of packets, as indicated by the traffic and connection parameters Each upstream time slot this counter is decreased by one, whereby, if the counter reaches a zero value, a pseudo-constant bit rate request is generated.

[0042] The jitter determining means JDM then receives the pseudo constant bit rate requests Rcbr and calculates therefrom the delay parameter J. A possible method may consist of assigning another so-called delay counter to each of the constant bit rate connections, entries in each counter indicating the time passed since the last grant for a particular connection was given. This time can be expressed in time-slot units. The contents of this counter then represents a measure for the age of this particular connection. By then comparing the contents of all these delay counters, the delay parameter J can be chosen as the maximum value of all entries.

[0043] The medium access controller MAC, adapted to perform the medium access control method of the present invention, calculates at each upstream timeslot a multi-permit. Such a multipermit consists of one or more network terminal grants, whereby each of the network terminal grants such as for instance G1 being a grant for network terminal T1, and Gi being a grant for network terminal Ti, includes one or more codes allocated to this specific network terminal. In FIG. 2 an example of such a multipermit generated at a particular timeslot thereby includes a network terminal grant G1 for terminal T1, this network terminal grant G1 itself including code C1 and code Cm allocated to T1, and a network terminal grant Gi for network terminal Ti, whereby Gi itself includes code Cj for allocation to network terminal Ti. In the embodiment of FIG. 2 these multipermits themselves are generated by one of a plurality of code allocation means. However also other embodiments of a medium access controller according to the invention exist, whereby only one algorithm is used for generating such a multipermit.

[0044] In the embodiment depicted in FIG. 2, 4 of such code allocation means are depicted and denoted CAM1, CAM2, CAM3 and CAM4. Which one of the four code allocation means will at a particular timeslot generate the multipermit, is thereby selected by the selection means SM. This selection means SM selects one of the code allocation means based on the values of the load parameter, the delay parameter and the set of two-dimensional constraints, thereby generating specific control signals to the respective code allocation means, indicative of their selection and consecutive triggering or not. The selection is performed first by comparing the delay parameter J with a predetermined threshold value denoted Jmax. Jmax itself may be a system parameter calculated based on the delay bounds of all delay-sensitive connections. One possibility may consist of selecting the most stringent value of all of these values as stored in the connection admission control memory CACM. This comparison takes place in a first comparing means denoted SM1 included within SM. If the delay parameter J exceeds Jmax, the fourth code allocation means is triggered, by means of the value of control signal c4 transmitted between SM1 and CAM4. In case the delay parameter is lower than or equal to Jmax, a second comparing means denoted SM2 is triggered by means of control signal c5 transmitted between SM1 and SM2. This second comparing means SM2 is adapted to receive the activity parameter L and to compare this with two predetermined threshold values denoted Tfair and Teff. In case L exceeds Teff, the first code allocation means CAM1 is triggered by means of control signal c1 generated by SM2 and transmitted to CAM1. In case the activity parameter L is smaller than Tfair, the second code allocation means CAM2 is triggered by means of a control signal c2 generated by SM2 and transmitted to CAM2. In case L lies between Tfair and Teff, the third code allocation means is triggered by means of control signal c3 generated by SM2 and transmitted to CAM3. The values of these predetermined threshold values Tfair and Teff themselves may be fixed pameters determined by the operator during initialization of the network and as such been stored within SM2. In another embodiment, such as depicted in FIG. 2, these may also be dynamically updated during the operation on the basis of for instance an average value of the activity parameter calculated so far, and on the basis of the set of two-dimensional physical constraints Phy generated earlier by the physical constraint means.

[0045] The calculation of the values of Teff and Tfair takes place in a calculation means, denoted CM, and also part of the selection means SM.

[0046] A possibility thereby consists of selecting Tfair and Teff as a predetermined percentage of the available upstream bandwidth, for instance 50% for Tfair and 300% for Teff of this upstream bandwidth of the common transmission channel.

[0047] The different code allocation procedures performed by the distinct code allocation means CAM1 to CAM4 will now be described.

[0048] The first code allocation means CAMP is adapted to generate multipermits such that the load within the network, and represented by the activity parameter L, is reduced as much as possible, while still taking into account the two-dimensional physical boundary conditions. The reason for this is that in this case the activity parameter, being a measure for the load, already exceeded predetermined threshold Teff, indicative of saturation within the network. In order to remedy this, the throughput is maximised by serving the terminals with the highest number of requests and by allocating as much as possible codes to these terminals. In an extreme case all codes are thereby allocated to one terminal for which most data packets are waiting. In another case, the available codes may be divided amongst the two most loaded terminal. Which one is to be used will depend on the physical boundary conditions. The first code allocation procedure is also called the efficient state algorithm. Input parameters for the first code allocation means CAM1 are thus the requests R1, . . . ,Ri to Rn, provided by the request table RT, and the set of two-dimensional constraints Phy provided by PCI.

[0049] The opposite case occurs when the load is lower than the other predetermined threshold Tfair. In this case fairness may become the primary goal, whereby as much as possible terminals, representing a first maximum amount of these network terminals, are served in a fair way, for instance by means of a round robin procedure. This algorithm thereby also maintains information about the terminals that were served during the preceding time slot, still taking into account the set of two-dimensional constraints Phy. This second code allocation procedure, performed by the second code allocation means CAM2, is therefore also called the fair state algorithm. Input parameters again consist of the requests from the request table and the set of two-dimensional constraints Phy.

[0050] In the case that the activity parameter L lies between the two predetermined thresholds Teff and Tfair, an appropriate equilibrium between fairness and efficiency is aimed at. The third code allocation procedure, performed by the third code allocation means CAM3 is thereby similar to the fair state or second code allocation procedure, but in stead of allocation only one code to one terminal at a time, a predetermined number of codes are allocated to one terminal. This predetermined number is denoted n and is as well calculated by the calculating means CM, for instance by dividing the number of codes by the number of terminals, in case the network is configured such that the result of this division is a number larger than one. n is provided by the calculating means CM to the third code allocation means CAM3. It is evident that in this case another, second, maximum amount of network terminals is assigned a code.

[0051] The second and third code allocation procedure can be further refined by taking into account traffic and connection parameters pertaining to individual connections requested by the terminals. These traffic and connection parameters are schematically denoted TCP1, . . . ,TCPi, to TCPn for respective terminals T1, . . . ,Ti to Tn. In case the connections requested by the individual terminals are characterised by these traffic and connection parameters, these traffic and connection parameters are to be considered as weights during the second and the third code allocation procedures. As is well known to a person skilled in the art, these traffic and connection parameters represent bandwidth guarantees.

[0052] The drawing in FIG. 2 thus includes for CAM2 and CAM3 two possible sub-code allocation means, respectively denoted cam2s and cam2q for CAM2, and cam3s and com3q, for CAM3. Cam2s, resp. Cam3s perform the “basic” second, resp. third code allocation procedures, in case the terminals are to transmit data packets for which no specific traffic and connection parameters are foreseen. The input parameters are then merely the requests, as provided by the request table, and the set of two-dimensional physical constraints.

[0053] In case traffic and connection parameters have to be taken into account, TCP1 to TCPn are additionally provided to com2q, resp. cam3q. Within CAM2 and CAM3 a discriminating means (not drawn on FIG. 2 in order not to overload the drawing) selecting one of the two sub-algorithms is present.

[0054] In other embodiments cam2s, resp. cam3s can be simply part of cam2q, resp. cam3q, by letting the weights to have the value 1. Again this is then controlled by means of the discriminating means.

[0055] The fourt code allocation means CAM4, adapted to perform the fourth code allocation procedure always has priority on the other three, by means of the fact that the delay parameter is always controlled first, and that, in case the latter exceeds the predetermined threshold value Jmax, the fourth code allocation procedure is always performed. This fourth code allocation procedure is therefore also called the exception state algorithm.

[0056] This exception state algorithm serves by priority the connections for which the delay bound risks to be broken, the so-called time-out connections. This algorithm aims at reducing the delay parameter J to a predetermined minimum value, in one embodiment corresponding to the predetermined threshold value Jmax. The working of the algorithm performed by CAM4 depends upon the number of time-out connections that must be served. This number is calculated ased upon the connection delay parameters, either directly provided by the connection admission control memory CACm, or indirectly based on the pseudo-requests calculated by the request table RT. The latter version is shown in FIG. 2, with the arrow with reference Rcbr representing the pseudo-requests as delivered from the request table.

[0057] Based on this number, and taking into account the set of two-dimensional constraints Phy, a maximum number q of codes that can be allocated to a connection is again calculated within CAM4. This is followed by serving the connections in a similar way as CAM3 did, but now with q in stead of n.

[0058] In FIG. 2 the third code allocation means CAM3 is selected for performing the calculating of the multipermit at a particular timeslot. The resulting multipermit is designated as {G1 (C1,Cm);Gi(Cj)} and thus includes, by way of example a first terminal grant G1 for terminal T1, with codes C1 and Cm assigned to this terminal, and a second terminal grant Gi for terminal Ti, with code Cj assigned to Ti.

[0059] At the terminal side, for instance network terminal Ti, the network terminal includes extraction means, adapted to extract from the incoming bitstream of multi-permits, the grants associated to this network terminal. Furthermore this extraction means has to extract from the network terminal grant, the code included therein. The network terminal also includes coding means which is adapted to encode data packets queued in the terminal, with the extracted code, before upstream transmission of the encoded data packets to the central station.

[0060] In a more sophisticated variant of the method, not only grants including codes per terminal are calculated, but the code itself is allocated to a storage queue within the network terminal. These storage queues within the terminals are shown on FIG. 3 and denoted TiQ1 to TiQk. These storage queues serve to temporarily store upstream data packets, in accordance to the their associated service categories. For ATM data packets these service categories are standardized by the ATM Forum specification AF-TM-0056.000 dated April 96. The codes calculated within the medium access controller are then not only allocated to a network terminal, but to one of the storage queues within the network terminal. The algorithm described in the previous paragraphs for calculating the multipermits basically remain the some, but are now based on requests per storage queues in stead of on requests per terminal. The set of two-dimensional constraints however remains at the terminal level, since of course no physical interference between the queues within one terminal is to be expected. The different code allocation means, receiving requests or pseudo-requests per service category and per terminal, in this case of course need to take this fact into account. A person skilled in the art is however adapted how to develop additional algorithm in order to group the requests per terminal, for taking into account the physical interference aspect, while further considering them apart in order to optimally divide the available codes per storage queue and per timeslot.

[0061] At the terminal side, the extraction means, denoted Emi in FIG. 3, was already adapted to extract from the bitstream of multipermits, the associated network terminal grant. In FIG. 3 such a multipermit is denoted as {Gl(C11,Cmk);Gi(Cj1)}, indicating that this multipermit includes a grant for network terminal 1, this grant including two codes: a first one for storage queue 1, with code C1, and a second one for storage queue k, with code Cm. Similarly, network terminal grant Gi for Ti, includes code Cj, for storage queue TiQ1. The extraction means Emi is than further adapted to extract this code and the associated storage queue from the network terminal grant. The extracted code Cj is consecutively transmitted to the coding means COMi, whereas a control signal s is transmitted to the selected storage queue TiQ1. Upon receipt of this control signal, data packets are transmitted from TiQ1 to the coding means COMi. The latter will encode these data packets with the code Cj, which are then upstream transmitted to the central station.

[0062] While the principles of the invention have been described above in connection with specific apparatus, it is to be clearly understood that this description is made only by way of example and not as a limitation on the scope of the invention, as defined in the appended claims.

Claims

1. (Cancelled).

2. The method according to claim 3, said method further comprises

determining a set of constraints from said plurality of codes and from the physical interferences between said network terminals and said individual network terminal channels of said communications network.

3. A method to assign upstream timeslots and codes to a network terminal of a plurality of network terminals in a communications network comprising a central station coupled to said plurality of network terminals via the cascade connection of a common transmission channel and respective individual network terminal channels said network terminals transmitting upstream data packets to said central station in a time multiplexed way over said common transmission channel using said upstream time slots, and said communications network further comprising a medium access controller, said method comprises:

assigning at least one code of a plurality of codes available within said communications network for encoding said upstream data packets to be transmitted by said network terminals for inclusion within a network terminal grant assigned to said network terminal,

determining an activity parameter related to the load within said network, said network terminal grant being determined based on said load,

generating a network terminal grant, wherein said network terminal grant is generated by said medium access controller and said network terminal grant assigns said upstream time slots to one of said network terminals, and

upon receipt of said network terminal grant, said network terminal will encode upstream data packets within said network terminal with said at least one code prior to upstream transmission of said upstream data packets of said network terminal to said central station.

4. The method according to claim 3, wherein

said activity parameter is derived from requests transmitted by said network terminals to said medium access controller, and indicative of the amount of upstream data packets said network terminals intend to transmit to said central station.

5. The method according to claim 3, said method further comprises

determining a delay parameter related to delay sensitive connections within said communications network, said network terminal grant thereby being determined based on said delay parameter.

6. The method according to claim 3, said method further comprises

selecting one of a plurality of code allocation procedures for the determination of said network terminal grant, the selection being based upon the value of said activity parameter and the value of a delay parameter and physical interferences between said network terminals and said individual network terminal channels of said communications network.

7. The method according to claim 6, wherein

said plurality of code allocation procedures comprises a procedure to reduce said load within said network.

8. The method according to claim 6, wherein

said plurality of code allocation procedures comprises a procedure for simultaneously allocating a respective one of said plurality of codes to a respective one of a maximum amount of said network terminals.

9. The method according to claim 6, wherein

said plurality of code allocation procedures comprises a procedure whereby a predetermined amount of respective ones of said codes are simultaneously allocated to a respective one of a maximum amount of said network terminals.

10. The method according to claim 6, wherein

said plurality of code allocation procedures comprises a procedure that reduces said delay parameter within said network to a predetermined minimum value.

11. The method according to claim 8, wherein

said procedure uses traffic and connection parameters related to said upstream data packets within said network terminals.

12. The method according to claim 9, wherein

said procedure uses traffic and connection parameters related to said upstream data packets within said network terminals.

13. The method according to claim 3, wherein

said method is performed at predetermined instances, and each of these instances results in the generation of a multipermit including one or more of said network terminal grants.

14. The method according to claim 3, wherein

said data packets within said network terminal are classified within said network terminal in accordance with their associated service category and temporarily stored in a plurality of storage queues, each storage queue being related to a respective one of said service categories, and

said at least one code is assigned to at least one of said storage queues, whereby, upon receipt of said network terminal grant, said network terminal will encode data packets from said at least one of said storage queues with said at least one code prior to upstream transmission of said data packets from said at least one of said storage queues to said central station.

15. (Cancelled).

16. The medium access controller according to claim 17, said medium access controller further comprising

physical constraint means adapted to determine a set of constraints from said plurality of codes and from input parameters related to the physical interference between said network terminals and said respective individual network terminal channels of said communications network.

17. A medium access controller for use in a communications network wherein a central station is coupled to a plurality of network terminals via a cascade connection of a common transmission channel and respective individual network terminal channels and said network terminals transmit upstream data packets to said central station in a time multiplexed way over said common transmission channel using upstream time slots, said medium access controller generating network terminal grants for assigning said upstream timeslots to said network terminals and to transmit said network terminal grants to said network terminals, wherein said medium access controller assigns at least one code of a plurality of codes available within said communications network for encoding upstream data packets to be transmitted by said network terminals to a network terminal and to include said at least one code into a network terminal grant assigned to said network terminal and generated by said medium access controller, and

said medium access controller comprising load monitoring means for determining an activity parameter related to the load within said network, said medium access controller generating said network terminal grant based upon said activity parameter.

18. The medium access controller according to claim 17, wherein

said load monitoring means receives requests transmitted by said plurality of network terminals to said medium access controller, said requests comprising information related to the amount of upstream data packets said network terminals intend to transmit to said central station, said load monitoring means being adapted to determine said activity parameter from said requests.

19. The medium access controller according to claim 17, said medium access controller further comprising

jitter determining means adapted to determine a delay parameter related to delay sensitive connections within said communications network, said medium access controller being adapted to generate said network terminal grant based on said delay parameter.

20. The medium access controller according to claim 17, said medium access control further comprising

a plurality of code allocation means and a selection means, said selection means determining which one of said plurality of code allocation means will be activated for determining said network terminal grant.

21. The medium access controller according to claim 20, wherein

said plurality of code allocation means comprises a code allocation means for allocating codes such that said load within said network is reduced to a minimum.

22. The medium access controller according to claim 20, wherein

said plurality of code allocation means comprises a code allocation means for simultaneously allocating one distinct respective code of said plurality of codes to distinct respective terminals of a maximum amount of said network terminals.

23. The medium access controller according to claim 20, wherein

said plurality of code allocation means comprises a code allocation means for simultaneously allocating a predetermined amount of distinct respective codes of said plurality of codes to distinct respective terminals of a maximum amount of said network terminals.

24. The medium access controller according to claim 20, wherein

said plurality of code allocation means comprises a code allocation means for allocating codes such that said delay parameter is reduced to a predetermined minimum value.

25. The medium access controller according to claim 22, wherein

said code allocation means allocates codes based on traffic and connection parameters related to said upstream data packets within said network terminals.

26. The medium access controller according to claim 23, wherein

said code allocation means allocates codes based on traffic and connection parameters related to said upstream data packets within said network terminals.

27. The medium access controller according to claim 17, wherein

said medium access controller is further adapted to generate, at predetermined instances, a multipermit including at least one of said network terminal grants.

28. The medium access controller according to claim 17, wherein

said medium access controller is further adapted to assign said at least one code to at least one storage queue of said network terminal.

29. A network terminal for use in a communications network wherein a central station is coupled to a plurality of network terminals including said network terminal, via the cascade connection of a common transmission channel and respective individual network terminal channels, said network terminal comprising:

extraction means for detecting an associated network terminal grant and extract at least one code within a downstream bitstream of network terminal grants transmitted from a medium access controller included in said communications network, to said network terminals, said network terminal transmitting upstream data packets to said central station upon detection of said associated network terminal grant,

said associated network terminal grant being based upon said activity parameter related to the load within said network, and

encoding means for encoding said upstream data packets with said at least one code prior to transmitting said upstream data packets to said central station.

30. The network terminal according to claim 29, said network terminal further comprising a

plurality of storage queues adapted to store said upstream data packets in accordance to their associated service category, wherein said at least one code is thereby assigned to at least one storage queue of said plurality of storage queues,

said extraction means determining from said at least one code said at least one storage queue to which said at least one code is assigned, and

said encoding means for encoding upstream data packets from said at least one storage queue prior to transmitting said upstream data packets from said at least one storage queue to said central station.