Method for monitoring a transmission system which comprises a plurality of virtual, asynchronously time-shared transmission channels via which a data flow can be transmitted
For monitoring a transmission system comprising a number of asynchronously operating channels through which data cells can be transmitted, variable values are stored which, on arrival of a data cell, are altered by an alteration algorithm in accordance with data cell traffic and are then processed by an evaluation algorithm to produce an evaluation signal. Such evaluation signals are processed by a combination algorithm to produce a control signal. A data cell is or is not, transmitted, depending on the value of the control signal. Quality classes are assigned to the data cells and for each quality class a first variable value is stored, which is altered by a first alteration algorithm when a data cell arrives at a first location and is processed by a first evaluation algorithm to produce a first evaluation signal. The channels are divided into groups, each comprising those channels which use a particular potential congestion point in the transmission system. For each group of channels a second variable value is stored, which is altered by a second alteration algorithm when a data cell arrives at a second location and is then processed by a second evaluation algorithm to produce a second evaluation signal. Depending on the value of the second evaluation signal and depending on the value of the respective quality class, the first alteration algorithm or the first evaluation algorithm or both are so controlled that the number of data cells transmitted is reduced and this the more for data cells of lower the quality classes.
A frame relay exchange apparatus which accommodates a plurality of subscriber lines and relay lines and transmits the frames received from these lines to a destination line, constituted so as to measure the amount of communication data for every PVC line along with an elapsed time and so as to calculate the limiting coefficient .alpha. varying in accordance with the degree of the queue length of the transmission queue. When the measured amount of communication data exceeds the upper limit value of the terms of subscription of communication of the respective subscribing PVC lines, that upper limit value is changed to a reference value limited by the limiting coefficient .alpha., and then a frame discard operation and advance notification are performed. By this, the PVC performing communication within the range of the terms of subscription of communication is protected from unjust limitation from another PVC performing communication exceeding this, while utilizing the resources inside the switching network to the maximum level.
Method for detecting data packets, such as ATM data cells. The preamble comprises a bit alignment word (AW) and a unique word (UW). Successively, the Hamming distances, hd's, are calculated from the first detected l-bit, over a length of AW plus UW, then from the next bit over a length of UW, and finally from the second 1-bit, again over a length of AW plus UW. The position of the first bit of the detected data packet is then calculated on the basis of the hd values thus calculated.
Reactive congestion control in an asynchronous transfer mode (ATM) network where the network is formed by the interconnection of nodes each including a forward path for transfer of information from source to destination through the network and a return path for returning congestion control signals. Each source includes a modifiable issue rate unit which issues forward information signals at different rates for virtual channels in response to the presence and absence of congestion signals received on the return path. When congestion occurs, for example, a queue exceeds a threshold, a request is made to send congestion signals, in the form of backward explicit congestion notification (BECN) cells, back to the sources of the virtual channels currently submitting traffic to the queue. On receipt of a BECN cell on a particular virtual channel, a source reduces it transmission rate for the indicated virtual channel. If no BECN cells are received on a particular virtual channel for a certain period of time, a source may gradually restore its transmission rate. In order to reduce the number of congestion cells, a filter is provided for filtering requests for congestion signals. Fairness is promoted in the recovery of transmission rates using a rate recovery algorithm that increases slower rate transmitters faster than higher rate transmitters.
The upper and lower bounds of the probability distributions of the number of arrivals at given arrival rates and the upper and lower bounds of the distributions of the arrival rates of each call class are prestored in memory, and by referencing such prestored values, the distributions of the arrival rate and the number of arrivals are obtained (step 104), then, the upper and lower bounds of the resulting cell length distribution are calculated to obtain safe-side evaluation values of QOS parameters (step 106). Based on the obtained QOS parameters, admission or rejection of a call is determined (step 108). Using this method, a function defining the upper bound of the probability distribution of the cell flow rate at the outgoing line side can also be obtained, from which a threshold value is calculated, and abnormality in the cell flow rate is judged on the basis of the threshold value. Furthermore, using a similar method to the above one, it is possible to evaluate the cell length distribution when a UPC is provided on the outgoing line side, and from the result of the evaluation, an optimum declared value to the UPC can be obtained to set up an internetwork connection via the UPC.
A probability transition matrix S.sub.t is formulated that expresses a multiplexing process which includes nested threshold priority control and classified priority control. A state equation using the matrix S.sub.t is solved by substituting therein upper and lower bounds of a probability distribution of a cell arrival count a.sub.t in an average time series, to calculate a probability distribution of a cell length in a buffer. From the thus calculated cell length probability distribution, the QOS is evaluated on a priority class basis. Based on the QOS evaluation, optimization of call admission control and priority control is accomplished.
