Home > other > Test: stand on the Internet architecture, how do you deal with?
Test: stand on the Internet architecture, how do you deal with?
Time:09-16
Something connected to the system, have A central machine, multiple terminals, terminal type A, B, C, D, E, A total of five types, the future will also add other type (A) the average new monthly in the end, they were carrying A, B, C, D, E, F, G, H, A total of eight kind of sensors, the future will also increase the sensor type new A sensor (half year), the other customer number, distribution of the whole of China, but different customers purchasing terminal number, the least only less than 10, the most in 1000, the number of procurement
A: 1 * A B: 1 + 1 * A * B C: B + 2 * 1 * C + 1 * D Dean: 2 * D + 1 + 1 * * E F E: 2 + 2 + 2 * G * F * H
The precision of each sensor below A - uint32 B - uint16 C - uint16 D - uint32 E - float32 F - float16 G - uint8 H - uint8
Terminal machine centre and the communication protocol design is as follows: Because statistics A, B, C, D, E, A cumulative sensor for 1, for 2 B and C for 1, for 2 D, E to 1 A, 2 F, G for 2, H 2, so the unified communication protocol frame format,
Each represents a byte [], uint32 use [i3] [i2] [i1] [i0] said, uint16 (ih) and [il], said float32 with [f3] [f2] [f1] (f0), said float16 expressed in [fh] and [fl]
[A i3] [i2] A [A i1] [A i0] [B1 ih] [B1 il] [B2 ih] [B2 il] ih [C] il [C] [D1 i3] [i2] D1/D1 i1/i0 D1/D2 i3/D2 i2/i1 D2/D2 i0 f3 [e] [e f2] [e] f1 f0 [e] [f1 fh] [f1 fl] [f2 fh] [f2 fl] [I] g1/g2 I [h1] I/h2 I
A frame sensor data, a total of 30 bytes, sensor types is not supported, the corresponding data fill 0,
Question:
1, this kind of communication protocol is what defect? Considering the cost of a product can't use the OTA
2, if you to maintain the system, please stand on the boss, the department manager, single-chip computer programmers, Web programmer's position, how to guarantee their interests?
CodePudding user response:
All into a floating point number or int32 processing