GB2190566A - Terminal apparatus - Google Patents

Abstract

To permit fault detection in terminal apparatus of a TDM network, echo channel bits (E-bits, Figure 1) are monitored to determine if an apparent collision with another terminal on the network has occurred. Each time data collision is indicated a counter (2, Figure 2) is incremented, the counter (2) being periodically decremented. If a predetermined count is reached indicating apparent collisions occurring at higher than expected rates a fault indication is given. This enables terminal apparatus on a TDM network to detect faults which mimic collisions.

Description

SPECIFICATION Terminal apparatus The present invention relates to terminal apparatus and more particularly to terminal apparatus for use on time division multiplexed digital transmission networks.

In certain types of network such as the network known as the Integrated System Digital Network (ISDN) subscriber access to the network is controlled by a particular bit in time division multiplexed (TDM) frames.

CCITT Recommendation 1.430, Red Book, defines certain properties of the ISDN subscriber access pattern and certain of those properties are mentioned hereinafter to assist understanding of the invention.

Of interest is the frame structure defined for ISDN which in both transmission directions (Network (NT) to terminal (TE) and TE to NT) has the bits grouped into frames of forty-eight bits. However, the frame structure for NT to TE frames is different to the frame structure for TE to NT frames and the respective structures are now considered with referenceto Figure 1.

In the terminal (TE) to network (NT) direction the frame has the following organisation Bit position Group 1 and 2 Framing signal with balance bit 3-11 B1-channel with balance bit (first octet) 12 and 13 D-channel bit with balance bit 14 and 15 Auxiliary framing with balance bit 16-24 B2-channel with balance bit (first octet) 25 and 26 D-channel bit with balance bit 27-35 B1-channel with balance bit (second octet) 36 and 37 D-channel bit with balance bit 3846 B2-channel with balance bit (second octet) 47 and 48 D-channel with balance bit Bit 48 (the last bit of the frame or L-bit) is used to DC balance the whole frame.

In the opposed direction (NT to TE) frames transmitted by the network contain "E" channel bits which are used to retransmit "D" channel bits from the terminals. The "D" echo bits ("E" bits) are used to control D channel access. The complete NT to TE frame is assembled as follows: Bit position Group 1 and 2 Framing signal with balance bit =10 B1-channel (first octet) 11 E, D-echo-channel bit 12 D-channel bit 13 bit A used, for activation 14 FA, auxiliaryframing bit 15 Nbit 16-23 B2-channel (first octet) 24 E, D-echo-channel bit 25 D-channel bit 26 S1, reservedforfuturestandardisation 27-34 B1-channel (second octet) 35 E, D-echo-channel bit 36 D-channel bit) 37 S2, reserved for future standardisation 3845 B2-channel (second octet) 46 E, D-echo-channel bit 47 D-channel bit 48 Frame balance bit The S1 and S2 bits (bits 26 and 37) have been allocated to future system enhancements and at the time of the invention are unallocated in ISDN and consequently are tied to binary "0". As with the TE to NT frame structure bit 48 (the "L"-Bit) is used to DC balance the complete frame.

The procedure for accessing the D-channel allows several terminals connected in multipoint configuration to gain orderly access to that channel depends on monitoring of the E bits of the frame. Point to point operating terminals also operate following the procedure.

If more than one terminal attempts to access the D-channel at the same time the procedure ensures that only one terminal will succeed in completing transmission of its information. The transmitted information depends on the use of layer two frames delimited by flags represented by the binary pattern 01111110 and using zero bit insertion elsewhere in the transmitted data to prevent flag imitation.

Any terminal or network in an idle condition ie. not having any layer two frames for transmission send binary "ones" on the D-channel. The network on receipt of D channel bits reflects to the terminal the binary value received in the next available E channel bit position. Thus a terminal in an active condition monitors the E channel bits counting the number of consecutive 'ones' in that channel. On detection of a zero the count is restarted until the number of 'ones' received equals or exceeds a value dependant on the priority of the data to be transmitted.

Two kinds of priority hereinafter (priority one and priority two) are defined, signalling data being priority one and other data priority two. Within priority one and priority two there are also two levels of priority (A and B) a terminal being assigned to the B level within its priority class once it has successfully transmitted a layer two frame. The terminal returns to the 'A' level within its class once all other terminals within that priority class having 'A' level status have had the opportunity to access the D-channel for transmission of a data frame.

In ISDN priority one 'A' level permits access when eight binary 'ones' have been detected on the E channel whilst 'B' level is permitted access after nine binary 'ones' have been so detected. For priority two the respective numbers are ten and eleven binary 'ones' received on the 'E' channel.

In both priorities once 'B' level access is permitted terminal on 'B' level status reset to 'A' level status.

Once a terminal has access to the 'D' channel and commences transmission of its layer two frame monitoring of the E bits continues. The terminal compares the E-bit with the transmitted D-bit and, provided the bits are of the same polarity continues transmission. If the received E-bit is of opposite polarity to the D-bit then the terminal ceases transmission and assumes that a "collision" with another terminal has occurred. The other terminal (or if more than one other terminal is transmitting one of the other terminals) will normally successfully transmit its frame.

One particular problem which is likely to arise is that of a fault occurring in the networks, terminal access apparatus, interconnecting buses or the terminal itself which could mimic collision. Present terminals would not react to such faults since they are designed to release the bus on detecting collisions and await the requisite number of binary 'ones' on the 'E'-channel then to make a further access attempt.

It is an object of the present invention to provide terminal apparatus capable of detecting faults of the kind which mimic transmission collisions.

According to the present invention there is provided terminal apparatus of the kind comprising frame assembly means for assembling frames of data for transmission in digital form in predetermined data bit positions of TDM network frames, monitoring means arranged to monitor predetermined echo bit positions of the TDM network frames, the monitoring means being arranged to monitorthe echo bit positions when the terminal apparatus requires access to determine whether the data bit positions are free and when the terminal apparatus has access to determine whether the echo bit equals the transmitted data bit and, if not, to indicate occurrence of an apparant data collision, the terminal apparatus further comprising counter means which is incremented each time the monitoring means indicates an apparent data collision and which is decremented at periodic intervals, and comparison means arranged to compare the accumulated count in the counter means with a predetermined maximum and, if the predetermined maximum is equalled or exceeded, to provide a signal indicative of an error whereby the occurrence of a predetermined plurality of apparent collisions within a predetermined period is indicated.

Facilities may be provided to compare the outputs of several such comparison means each in a respective terminal of the kind defined in the preceding terminal whereby an apparent fault may be determined either as local to the terminal or as in common equipment.

An error detection circuit for use in a terminal in accordance with the invention will now be described by way of example only with reference to Figure 2 of the accompanying drawings which is a block diagram of the circuit.

Referring to Figure 2, the circuit comprises an Exclusive-OR logic gate 1, a counter 2, a decoder 3 and a timer 4. The counter 2 and the timer 4 receive clock signals from a clock synchronised to the E-bits of the forty-eight bit network TDM frame.

The last transmitted D-bit and the E-bit from buffer stores (not shown) are applied to the input of the EX-OR gate 1 which provides a signal to the up-count input of the counter 2 if the two signals are of opposite polarity.

Thus when the counter 2 is clocked it will be incremented if an apparent collision has occurred.

The timer 4 is programmable to provide an output signal after a number of clock pulses, the output signal being provided to the down-count input of the counter 2. Thus, periodically, the counter 2 will be decremented thus providing a counter of the kind sometimes cailed a "leaky bucket" counter.

The decoder 3 monitors the count held in the counter 2 and, if the count reaches a number preprogrammed in the decoder provides an output signal indicative of a fault.

It will be appreciated that the leakage rate programmed in the time 4 will depend upon system or network parameters as will be the value which triggers an output signal from the decoder 3.

However, in ISDN the probability of a collision is low and typical timer 4 setting may be in the order of one or two seconds with the decoder 3 set to output if two collisions occur in the one or two second period.

The output from the decoder 3 may be used to provide an audio, visual or audiolvisual indication of a malfunction to a terminal user.

It will be realized that the counter 2 should be inhibited from wrap-around, ie. should be inhibited from further decrementing when the count equals zero.

Whilst as hereinbefore described the circuit increments on detection of a fault and decrements at periodic intervals it should be noted that the circuit may be arranged to act in the reverse manner provided that the decoder 3 is then arranged to indicate a fault condition on a predetermined minimum value of the count rather than on predetermined maximum value.

Since terminals for use on ESDN incorporate an "S-bus chip" including circuiting for controlling bus access, and incorporating collision detection therein the additional circuit of the present invention could be added to the same integrated circuit. Alternatively the circuit of Figure 2 could be incorporated in an independant integrated circuit for separate inclusion in the terminal apparatus.

Comparison of the outputs of the decoders 3 in several terminals may be made such that if more than one terminal is showing a significant collision pattern, the fault may be determined as network rather than terminal related.

Claims (3)

1. Terminal apparatus of the kind comprising frame assembly means for assembling frames of data for transmission in digital form in predetermined data bit positions of TDM network frames, the monitoring means being arranged to monitor predetermined echo bit positions of the TDM network frames, the monitoring means being arranged to monitor the echo bit positions when the terminal apparatus requires access to determine whether the data bit positions are free and when the terminal apparatus has access to determine whether the echo bit equals the transmitted data bit and, if not, to indicate occurrence of an apparent data collision, the terminal apparatus further comprising counter means which is incremented each time the monitoring means indicates an apparent data collision and which is decremented at periodic intervals, and comparison means arranged to compare the accumulated count in the counter means with a predetermined maximum and, if the predetermined maximum is equalled or exceeded, to provide a signal indicative of an error whereby the occurrence of a predetermined plurality of apparent collisions within a predetermined period is indicated.

2. A network including a plurality of terminals each in accordance with claim 1 wherein outputs of more than one terminal are compared to enable determination of an apparent fault as being present either in a specific terminal or in equipment common to the plurality of terminals.

3. Terminal apparatus substantially as hereinbefore described with reference to the accompanying drawings.