As mentioned in a previous post, my RIT SE Senior Project team, the Crash Test Buddies, created an open-source library to handle Wi-Fi Direct on Android. This WiFi-Buddy library was used in several apps, one of which tested the Service Discovery between phones. From our experience and research, Service Discovery can be very unstable on Android. Sometimes phones discover each other nearly instantly, and other times they take much longer to discover. We also noticed that some phones never successfully discover each other. The purpose of this Wi-Fi Direct testing and data collection app is to determine what factors affect the Service Discovery between phones.
Since Wi-Fi Direct can be very inconsistent, we tried to control the test scenarios and limit as many variables as possible to produce consistent results. For all of the scenarios, two phones were placed about 50 feet apart in an open environment. One phone ran the wifi-direct-tester app to broadcast a local P2P service, and the other phone ran the DataCollectionTests app to continuously discover, unregister P2P, and rediscover that service. The discovering phone was rebooted after every 100 test cycles in order to help prevent any degradation. Both of the apps used for testing utilized WiFi-Buddy to handle the Wi-Fi Direct functionality.
In the first test scenario, a BLU Advance 5.0 phone discovered a similar BLU Advance 5.0, in an environment without any nearby Wi-fi networks. This was supposed to be the ideal scenario, testing with identical phones in an environment with little nearby network activity. Using the timings from the testing and data collection app, we calculated statistics and generated graphs to observe patterns. For this scenario, the mean time to discover a service was about 0.4 seconds. This would possibly be a reasonable discovery time for a crash avoidance application, but there were also many spikes in time. The maximum time to discover a service was about 28 seconds, and there were several timings around 5 or 6 seconds. This showed that even though the same test was being run continuously, in a fairly constant environment, there were intermittent issues and varying results. This also showed that even in the optimal situation, the time to discover a service is likely to occasionally be too long for a crash avoidance application.
In another test scenario, Service Discovery was tested with the same two BLU phones in an environment with many Wi-Fi networks nearby. The purpose of this test was to determine if the presence of network activity has a negative effect on Wi-Fi Direct Service Discovery. From experience, we were fairly confident that this would be true. The results from the data collection app showed that the mean time to discover the other phone was about 0.7 seconds, which was nearly double the time from the first scenario. There were still many spikes throughout the graph. From these results, we concluded that the presence of Wi-Fi networks likely has a negative effect on Wi-Fi Direct Service Discovery.
In a third test scenario, an LGE Nexus 5 was used to discover a BLU phone, in the same network-heavy environment used previously. The purpose of this test scenario was to determine whether or not discovering s different brand phone has a negative effect on Wi-Fi Direct Service Discovery. From our experience, we figured this would be true, because we often had trouble discovering BLU phones from Samsung phones. The testing results showed that after 1,000 cycles, the mean time for the LGE Nexus to discover the BLU was about 1.7 seconds. From these results, we concluded that the type of phone being used can have a negative effect on Wi-Fi Direct Service Discovery.
The testing results showed that even in the optimal situation, using identical phones in an open environment, the time to discover a service is usually quick, but can often reach several seconds. If the phones are discovering each other in an environment with many networks, the time to discover will likely increase. If two different phones are discovering each other, then the time to discover will likely take even longer. After Service Discovery, the phones still need to connect to each other, open sockets, and transfer data. The connection times vary as well, and some phones have trouble connecting to each other. Considering these results, we believe that there would need to be improvements made to the Wi-Fi Direct API by Android, and possibly improvements by phone manufacturers in order for Wi-Fi Direct to produce consistent, reliable results.
Our DMT carts come filled with 250mg, 500mg with 1ml,5ml of N, N-Dimethyltryptamine (DMT), which is sufficient for about 8 to 12 breakthrough experiences 30mg to 70mg. Our DMT is extracted from Brazilian Mimosa Hostilis root bark and re-crystallized twice for purity. Flavoring has also been added for an ultimate smooth inhale. Just select the flavor you like best!This method is safer and more convenient than using large or tricky apparatuses that require lighter which destroy DMT if touched by the flame. It also allows you to regulate your experience with each pull. Our DMT does not contain any vitamin e acetate.DOSAGE GUIDELINES: 4mg DMT per 3 seconds pullLight: 1-5 pulls 4 – 20 mgMedium: 6-10 pulls 24 – 40 mgStrong: 10+ pulls 40+ mgIDEAL SETTINGOur DMT is strong so only take in the safety and comfort of your own home. Sense of loss from reality is common, and complete audio and visual hallucinations can take place.Note:DMT tends to crystallize very easily and becomes solidified, especially during colder temperatures. In such cases, just run the cartridge under hot water for a few minutes or take a hair dryer or lighter to it.The provided vape pen may need charging (via the included USB charger) before use.Storage Directions: Keep in a dark & cool place. It is best to keep refrigerated.