Case Studies of Problems in Digital Cordless Phones

2.1 Overview and report

A customer using digital cordless phones (PHS: 1.9-GHz band) connected to NTT’s αNX II business phone system reported a problem in which calls could not be made or received within the customer’s retail store (behind the counter on the first floor: see Fig. 1). In this configuration, one cell station (CS) is located near the window on the second floor, and two PHS terminals are in use. Replacing the terminals did not solve the problem.

Fig. 1. Equipment configuration (customer’s store).

2.2 Results of measurement and cause of failure

A spectrum analyzer (Keysight Technologies N9020A) was used to measure the signal strength at the center of the store on the second floor and behind the counter on the first floor. The CS is located near the center of the store on the second floor, so we were able to observe signals that had sufficient strength (–40 dBm) at that measurement point (Fig. 2(a)).

DECT signals were also observed in addition to PHS signals. At the measurement point behind the counter on the first floor, the signal strength from the CS installed on the second floor was found to have dropped to as low as –70 dBm (Fig. 2(b)).

Fig. 2. Results of measuring signal strength.

The results of the above measurements led us to infer that this problem occurred because the PHS terminals were being used at a location that was too far away from where the CS was installed, resulting in insufficient signal strength at the terminal usage location.

2.3 Countermeasure

We received the customer’s consent to add a CS near the location where the PHS terminals were being used (near the first floor counter). We found that stable communications could be achieved with the signal strength of –65 dBm at the terminal usage location.

3.1 Overview and report

A customer using a private branch exchange (PBX) (EP74) reported noise and dropped calls when using digital cordless phones (PHS terminals) accommodated by the PBX. A total of 10 CSs were installed throughout the floor, and 109 PHS terminals were being used (Fig. 3). Replacing terminals failed to solve the problem.

Fig. 3. Floor diagram.

3.2 Results of measurement and cause of failure

We used a spectrum analyzer (Keysight Technologies N9020A) to measure the radio signal environment in the PHS band below CS 3 (see Fig. 3). In this measurement, DECT signals could be observed in addition to PHS control-channel and call-channel signals (Fig. 4).

Fig. 4. Radio signal environment (measurement point A).

The results revealed that the DECT and PHS signals were interfering with each other. This led us to infer that the cause of noise in the phones was this interference between DECT and PHS signals. Although PHS and DECT have functions for mutually avoiding interference, it can be inferred that a high channel utilization rate may leave no empty channel available for interference avoidance, resulting in the occurrence of interference and dropped calls.

3.3 Countermeasure

To prevent this phone noise from occurring, we recommended to the customer that no DECT terminal be used or alternatively that the DECT terminal be exchanged with a terminal using frequency bands other than the 1.9-GHz band (2.4-GHz digital cordless phones, 5-GHz Wi-Fi devices, etc.).

4.1 Overview and report

Dropped calls and line disconnections were occurring in digital cordless phones (PHS: 1.9-GHz band) accommodated by NTT’s αNX II-L business phone system. The customer reported that such problems occurred frequently in the processing room of the food plant depicted in Fig. 5. Measures such as replacing terminals or adjusting signal power failed to solve the problem.

Fig. 5. Equipment configuration (2nd floor of plant); PBX is installed in main distribution frame room (MDF) on third floor.

4.2 Results of measurement and cause of failure

We used a spectrum analyzer (Tektronix RSA6114A) to measure the radio signal environment at the measurement point within the processing room (see Fig. 5). The PHS signal strength was found to be sufficiently strong, and no interfering signals were observed (Fig. 6).

Fig. 6. Results of measuring signal strength.

The processing room has wall and floor surfaces and machinery made of stainless steel and other types of metal, so we considered that reflected signals might have an effect on call quality. We measured such reflected signals with a network analyzer (Keysight Technologies E5071C) and observed the delay (reflected) signals of nearly the same strength as direct waves and long-delay (reflected) signals of 80 ns or longer (Fig. 7). When compared with a location enabling normal communications (Fig. 8), the above results revealed the existence of a multipath.

Fig. 7. Results of measuring reflected signals (at time of a dropped call).

Fig. 8. Results of measuring reflected signals (normal communication).

The results led us to infer that dropped calls in the processing room were caused by reflected signals.

4.3 Countermeasure

The customer’s processing room with its wall and floor surfaces and machinery made of stainless steel and other metals is conducive to the generation of reflected waves, resulting in a critical environment for radio signals. We explained to the customer that such an environment made the use of digital cordless phones difficult and proposed the use of fixed-line phones. To prevent future problems, we recommended that call tests be performed before introducing a new service to see whether dropped calls or other problems might occur.