Through practice, organization, and analysis, I have summarized some embedded rules learned during the embedded development process for your reference:
Embedded computing not only requires fast and consistent calculations over a network but also demands that the system orderly executes its code and data stored in a “common” and “narrow” space.
Embedded computing requires not only rapid and effective calculations but also insists that the system continues to operate normally even when certain computational units fail.
The calculation results of embedded systems depend not only on the correctness of logical operations but also on the computation time of these results.
Once an embedded system has sufficient redundancy, the system’s “initial sensitivity” to its “final computation results” becomes negligible.
For embedded systems, the trend of structural complexity indicates: a simpler system structure is more effective (The simplest is the best); b a more complex system structure is more stable (More complex is more stable).
Simplifying an embedded system strengthens the interaction between the remaining functions; once the system’s functions are simplified, the probability of successful intrusion by external invaders increases.
If a certain function is to be retained in an embedded system, it is best to regard all other functions as “sacrosanct”; the removal (extinction) or generation (invasion) of system functions will significantly alter the overall (cluster) structure and its dynamic performance.
The most important aspect of the embedded Internet is often not the characteristics of individual devices within the network but the overall order present in the network, i.e., network order. In a highly redundant network, the individual role of devices no longer significantly affects the overall system performance, but rather the characteristics of all nodes and their connections play the main role.
A network composed of embedded Internet nodes interacting with each other often exhibits properties that are not significantly related to the properties of individual nodes.
Guaranteeing message searches: they have strict time sensitivity or basic normal operational requirements, and such messages require a time guarantee from the system. Once activities or tasks triggered by such messages are executed, their time constraints will be guaranteed by the system within a specified time interval. Optimal effect messages: they possess typical soft time constraints, meaning their time constraints are dictated by the time sequence of the activities or tasks themselves, and do not require system guarantees to meet their time constraints.
The immune system of the embedded Internet should emulate a living organism mechanism, where the immune function is a “feed-forward” system requiring the system to have foresight, allowing it to “attack (small) toxins with (large) toxins.”
The embedded Internet is a complex network that can be analyzed by simple “components” to interpret the “whole” or express the “global” with “nodes.”
If system A is an embedded system of system B, i.e., B(a), then the cost of system A should not exceed 10% of the cost of system B, while the cost of system B(a) should be greater than the sum of the costs of systems A and B, and the cost-effectiveness of system B(a) should increase by 30%.
Scan the QR code below to follow!
On the electronic road, let’s walk together!